Video Generation

Today's Stories

1. 1

   OpenAI shutters ChatGPT Atlas browser after less than a year

   OpenAI is shutting down ChatGPT Atlas, its browser launched in October, with deprecation targeted for August 9th. The company is rolling the lessons learned from Atlas into updated browser features in the desktop ChatGPT app and a new cloud browser for work mode. Atlas was designed to help users accomplish tasks on the open web on their behalf. OpenAI is consolidating what it learned from the experiment into more integrated productivity tools as it works to reduce side projects and catch up with Anthropic on work-focused features.

   The August 9th sunset date for Atlas marks one of several recent product shutdowns by OpenAI, including the video generation app Sora and a paused ChatGPT "adult mode," suggesting the company is refocusing its priorities.

2. 2

   Wistron chief: AI demand broadening beyond US as sovereign AI expands

   Wistron chairman Simon Lin stated that AI demand is expanding beyond the US as sovereign AI projects gain traction in more countries, and that the industry is entering a new phase rather than a bubble. The shift toward sovereign AI and broader geographic adoption suggests a longer runway for AI infrastructure spending globally, moving beyond concentration in a single market and opening new revenue streams as more regions deploy their own AI systems.

   The pace at which sovereign AI projects translate into paid services and sustained infrastructure investment across different regions will determine whether this expansion sustains the current momentum in the AI hardware and services sectors.

3. 3

   Apple Research Tackles Text-to-Video Audio Sync Challenge

   Apple researchers published a study addressing two key obstacles in text-to-sounding-video generation—the process of creating videos with synchronized audio from text descriptions. The work identifies that shared captions between text and audio create interference, and that a gap exists between detailed training captions and shorter user prompts at deployment. Generating video with perfectly matched sound is a foundational problem for AI video tools; solving it could improve how AI systems create multimedia content that feels natural and cohesive. The research proposes solutions to both the conditioning bottleneck and cross-modal feature interaction, potentially making video-generation systems more reliable for real-world use.

   The full research is available on Apple's machine learning website, offering detailed technical methods and findings for developers and researchers working on audio-visual AI systems.

4. 4

   OpenAI shuts Sora video app, Disney deal scrapped

   OpenAI announced Tuesday it is shutting down the Sora app, the video generation tool the company launched in December 2024. The $1 billion（約1600億円） partnership deal with Disney, announced in December 2025 to let users create videos with Disney characters, is also being terminated. OpenAI did not confirm reports that Sora would move into ChatGPT, but promised to share timelines for the app and API soon. Sora faced steep competition from other AI video tools like Google's Veo and Luma Ray and never captured the early momentum from its launch. The sudden exit and Disney deal cancellation, just months after CEO Bob Iger was promoting the partnership in February 2026, signal internal turmoil at OpenAI—a company also dealing with wrongful death lawsuits tied to teen mental health concerns.

   OpenAI has promised to share more details soon, including timelines for preserving user work. The deal was originally set to last three years; Disney said it "respects" OpenAI's decision to exit video generation and shift priorities elsewhere.